Surge Voltages in Residential and Industrial Power Circuits

Special instrumentation was developed for monitoring the magnitude and frequency of occurrence of surge voltages in residential and industrial circuits. Over a period of 2 years, more than 400 locations in 20 cities were surveyed. Monitoring was accomplished by automatic recording cathode-ray oscilloscopes and simple surge counters. In residential circuits, two significant sources of surge voltages were identified : load switching within the house, and external surges, most likely associated with lightning, coming through the service drop. In industrial circuits, the levels of surges are lower than in residential circuits. However, switching surges on the load side of the switch can be severe, Internally generated surges a s high as 2500 volts were recorded during fhis test program, and surges due to lightning reaching 5600 volts have been recorded on a 120-volt overhead distribution line. Paper 69 TP 618-PWR, recommended and approved by the Surge Protective Devices Committee of the IEEE Power Group for presentation at the IEEE Summer Power Meeting, Dallas, Tex., June 22-27, 1969. Manuscript submitted February 17, 1969; made available for prmting April 7, 1969. The authors are w ~ t h the Research and Development Center, General Electric Company, Schenectady, N. Y. 12305. T HE succeesful operation of semiconductors and new insulation systems in appliances and consumer electronic devices may be adversely affected by transient overvoltages occurring on the 120-volt power supply. Detection and measure ment of these transient surges permits a designer to provide suitable bu ih in tolerance, or a t least to recognize that there is a need for appropriate suppression or protection. Special instrumentation has been developed for detection and measurement of surge voltages. I n the first phase of testing, automatic recording cathode-ray oscilloscopes (CRO) provided complete information on waveshape, and on magnitude and frequency of occurrence at a small number of arbitrarily chosen locations. These data, although statistically restricted, were adequate to demonstrate the existence of the problem, define some sources, and indicate typical waveshapes. I t also became clear that there was a need for a larger sample that would be more statistically valid. In the second phase of testing, 100 surge counters, simple in design and easy to install, were developed especially for this program. These counters were installed in several hundred homes in various cities located in the East and Midwest of the United States. The purpose of this paper is to discuss recordings obtained during the two phases of this program and the statistical aspects of the results. I t is hoped that publication of this paper will encourage other investigators to publish their findings. This would provide a broader statistical base for verification of the frequency and magnitude of surge voltages. I n regard to industrial circuits, less data have been accumulated, but a number of significant case histories are presented to illustrate the problems likely to be encountered in that field.